[theonewithin]

So before I go out and buy/build something I thought I had better check my thinking.

Injector dead time. And I right in that you can use an oscilloscope and current clamp to get this info?

I know that you can see when the coil is charging and when it actually opens from watching the current ramp etc but I don't know if this is actually accurate enough?

Hoping David could chime in

Cheers.

[turbocg3]

I have wondered about this also. My injectors have an opening time of 1.3ms and a closing time of .7ms. I have tried both 1.3ms and .7ms. I use .7ms because AFR seem to be more consistent. But if you click the little icon it says the net opening and closing. Which would be 2.0ms in my case. I have not tried 2.0ms.

BTW I have decapped gm injectors from a junkyard 5.3l and my flow is set to 700cc. Gamma correction shows 115-130%.

[PSIG]

Considering there are several small variables that can influence results (and why it's a good idea to test for correct latency on the running engine), it can get you close for a given test voltage. Using @turbocg3's example numbers, and assuming good measurements; opening delay is no-flow, and closing delay is extra flow. So for his I would use [1.3 - 0.7 = 0.6ms] net dead time as a starting point.

One up-side to current testing is that you can use different voltages to get your correction curve. Even if the base value is off a bit, because Speeduino uses % voltage corrections, the curve of test % corrections will still be good. Be sure to test with correct fuel pressure differential, as that's one of those little controllable factors that have influence on your results, and why increasing fuel pressure does not predict increased fuel flow (higher pressure should be more flow, but also causes greater latency, so…). Uncontrollable or unknown factors include pintle bounce, temperature, control circuit efficiency, and so on. Does that help?

David

[theonewithin]

It does indeed!

[Damouk]

I have been battling with inconsistent afrs at idle and eventually narrowed it down to the battery voltage swinging about (another story), but what I have also done is create a map to allow then deadtime to be 'checked' at a range of voltages. The tps is used to vary the injector duty around the dead time. With this you can then check when the injector starts to squirt fuel.
I know this is a crude way of doing things but it worked very well.

[PSIG]

@Damouk - you have my interest, but I'm not following all of your scheme. Can you explain the basic strategy?

David

[gpineau]

PSIG, I think he is doing something similar to the tester I made. Pulse the injector with increasing pulse duration until it starts to "leak" more fuel.

Remember you taught me ....Dead Time = (StaticFlow - PulsedFlow) / (StaticFlow / Milliseconds) / NumberOfCycles

Where:
Dead Time includes both opening and closing delay (lag) effects;
StaticFlow is the open-flow (non-pulsed) test volume;
PulseFlow is the pulsed (cycled) flow volume of identical ON time;
Milliseconds is the total injector ON test time per test (e.g., 10 seconds = 10000 MS);
NumberOfCycles is the number of pulses in the cycling test...

I call it Dave's formula. And it worked well for me.